Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
01-05-2015, 01:08 PM   #16
Pentaxian
Pablom's Avatar

Join Date: Feb 2010
Location: Jerusalem
Posts: 1,940
I believe the new Nikon D810 has ISO settings as low as 32, that is a great feature just for the practicality, even if it gains nothing in the form of IQ.

Since there are films with as low iso as 1 I don't see why it wouldn't be possible to achieve the same with a digital sensor. A reason I might not see it is that I am clueless about the technologies behind both digital sensors and photographic film.

01-05-2015, 01:21 PM   #17
Site Supporter




Join Date: Sep 2011
Location: Richland, Washington, USA
Posts: 879
So sounds like the best place to shoot is iso 100, reserving knocking down to 80 for situations where your lighting/aperture/shutter speed needs require a lower sensitivity.
01-05-2015, 01:51 PM   #18
Forum Member




Join Date: Feb 2009
Location: Kentucky
Posts: 91
Original Poster
Well that's a lot of interesting info... Thanks to all for responding... I wonder what the actual normal ISO would equate to. IE it's probably like 84 or 123 and that is what pentax calls 100 and goes from there...
01-05-2015, 01:55 PM   #19
Pentaxian




Join Date: Dec 2010
Location: Ontario
Photos: Gallery
Posts: 2,567
QuoteOriginally posted by jrpower10 Quote
So sounds like the best place to shoot is iso 100, reserving knocking down to 80 for situations where your lighting/aperture/shutter speed needs require a lower sensitivity.
Nope, iso 80 wins for quality in measurable ways, check out Tests and reviews for the camera Pentax K-5 Measurements - DxOMark (similar results for the k5ii/s). The difference to iso100 is of course small, but it's there.

Not all cameras 'expanded' low iso's offer an increase like this. I dunno if it was software voodoo magic pentax used or some hardware thing pentax was able to exploit.

---------- Post added 01-05-15 at 03:58 PM ----------

QuoteOriginally posted by soycory Quote
Well that's a lot of interesting info... Thanks to all for responding... I wonder what the actual normal ISO would equate to. IE it's probably like 84 or 123 and that is what pentax calls 100 and goes from there...
On the k5, iso 100 is really iso 91, at least according to the dxo tests. This is an area where I think Pentax does pretty well relative to other manufacturers, their stated iso's are close to reality.

01-05-2015, 02:21 PM   #20
Pentaxian
emalvick's Avatar

Join Date: Feb 2008
Location: Davis, CA
Photos: Gallery
Posts: 1,475
As others have pointed out, the native ISO is related to the sensor, but one can't neglect the affect the exposure level has to define that specific native ISO. dSLRs and digital camera's in general have always tried to make things comparable to the film world, so the native ISO is the number that allows the sensor to capture exposures photographers can comprehend under specific conditions they're used to.

I do think that it might be arbitrary that most companies have developed their sensors to provide native ISO's around 100. I'm not sure there would be huge benefits to lowering the native ISO, although it could help reduce the need for things like ND filters if the ISO could get low enough. Based on an earlier post, I suppose the multiple-exposure functions could be used to emulate a lower ISO, although I think it is more of an emulation of a lower shutter speed or ND filter or perhaps a combo of all the above.

The ISO 80 for the Pentax may have just been a software trick or perhaps hardware; could they drop the voltage or current to the sensor and achieve something like a low ISO?
01-05-2015, 02:32 PM   #21
Senior Member




Join Date: Feb 2014
Photos: Albums
Posts: 132
Just so happened.. Actually SONY cameras I've owned will notify you with different color on display if you are not on the true native ISO. For example the Sony A7. Native ISO is 100. So anything under ISO100 is possible, but you won't get the best result and they were marked under orange color with underline instead of white color.

Camera will auto apply the noise reduction when noise ratio is getting too high on high ISO. Similar to lower ISO. If A7 you are not shooting on ISO100, you will see the light source breaks up into ugly pixels when set to ISO50 or ISO64.. And that's the same effect to high ISO such as 12800. But this ugly light source pixels won't be there around ISO100~ISO800 or even ISO1600..

It's good that SONY did indicated that anything under ISO100 is actually "simulated ISO". At least they are honest about it.
01-05-2015, 02:42 PM   #22
Site Supporter
RGlasel's Avatar

Join Date: Sep 2013
Location: Saskatoon
Photos: Gallery | Albums
Posts: 1,345
QuoteOriginally posted by vonBaloney Quote
the buckets (pixels) can only collect so many photons. This is a hardware property. The size of a full bucket corresponds to the native ISO of the sensor system. If the native ISO were "only theoretical", well then we'd be able to change the native ISO with a dial on our cameras, and noise at 800 would be just as low more-or-less than at 100 because we'd have changed the "real ISO" of the whole system somehow. But hardware is hardware -- there are very real physical limits.
Actually, the sensor has a bias voltage applied to its circuits (think of it as a flatline signal) and variations in the output signal compared to the bias voltage is the signal from the sensor. Photon buckets is not a useful analogy, if enough photons hit the photosite or pixel to change the voltage flowing through the pixel to be detectable, the difference between the output voltage and the input voltage is the signal. If the bucket doesn't get full enough to produce a detectable change in voltage, there is no signal and if the bucket is overwhelmed with light energy, the circuitry in the sensor can't distinguish how "overfull" the bucket is, either.

The K-30 has an EV range of 0 to +22, the K-3 is even better. 0 EV corresponds to an illuminance value of 2.5 lux or 0.23 footcandles and 22 EV is 10,496,000 lux or about 975,000 footcandles. That's more than a 4 million fold range in illuminance (the human eye can perceive approximately a 2 TRILLION fold range of light intensity). A piece of metal oxide only a few micrometers across can't handle that range of light intensity without the camera electronics controlling the gain or bias voltage applied to the sensor. So if the camera is adjusting the sensitivity of the sensor (based on a signal from the metering system), how can the sensor have a native ISO?
01-05-2015, 07:09 PM   #23
Pentaxian




Join Date: Oct 2011
Location: Savannah, GA
Posts: 5,002
QuoteOriginally posted by RGlasel Quote
Actually, the sensor has a bias voltage applied to its circuits (think of it as a flatline signal) and variations in the output signal compared to the bias voltage is the signal from the sensor. Photon buckets is not a useful analogy, if enough photons hit the photosite or pixel to change the voltage flowing through the pixel to be detectable, the difference between the output voltage and the input voltage is the signal. If the bucket doesn't get full enough to produce a detectable change in voltage, there is no signal and if the bucket is overwhelmed with light energy, the circuitry in the sensor can't distinguish how "overfull" the bucket is, either.
Hmmm...that's why it is a useful analogy -- because like a bucket once it is full you can't tell anything but "at least this much" (the capacity of the bucket) was there to collect. How much more was there? We don't know -- it ran out of the bucket. The analogy is showing the crudeness of the sensor, not its sophistication. I wish an analogy of [some more precise and much less limiting counting device] was apt, but the lowly bucket is closer.

QuoteQuote:
The K-30 has an EV range of 0 to +22, the K-3 is even better. 0 EV corresponds to an illuminance value of 2.5 lux or 0.23 footcandles and 22 EV is 10,496,000 lux or about 975,000 footcandles. That's more than a 4 million fold range in illuminance (the human eye can perceive approximately a 2 TRILLION fold range of light intensity). A piece of metal oxide only a few micrometers across can't handle that range of light intensity without the camera electronics controlling the gain or bias voltage applied to the sensor. So if the camera is adjusting the sensitivity of the sensor (based on a signal from the metering system), how can the sensor have a native ISO?
How can it get a signal from the metering system to tell how many photons are going to arrive if they haven't arrived yet? It gets a signal from the camera operator as to what ISO to set it at, which means what gain/voltage. It sounds like you are saying the sensor automatically turns up the gain if the scene is dim without you telling it to (by setting the ISO). If that were true, you could trick the sensor by shooting your flash (without warning), and it wouldn't know what to do with bulb mode at all. The sensor isn't controlled by the metering system. (Other than in a programmatic way with an automatic mode where the metering acts as an input to the exposure calcuation. And then the camera mode or the camera operator has the option to change the ISO to get a "correct" exposure.)

If there was no native ISO, then all ISOs would be the same (or could be the same), and you'd have equal noise across the range because the higher ISOs would not need (more) boosting (which increases noise). So if you want to forget about the exact technical details, native ISO is simply that point with the cleanest images. I understand there is some slop in there, and it isn't necessarily an exact number, but nevertheless the concept is valid.

01-05-2015, 07:48 PM   #24
Senior Member
Omestes's Avatar

Join Date: Nov 2011
Location: Phoenix, AZ
Photos: Gallery | Albums
Posts: 207
QuoteOriginally posted by ScooterMaxi Jim Quote
Bottom line, Pentax users should be setting HC (at least in Auto to avoid blown highlights) to get the most out of their cameras - and match or even exceed the performance built into the Nikon equivalents.
Is there anything to back this up? I generally keep all the correction settings off, hoping for a more unprocessed image. What exactly does highlight correction do, either hardware or software wise?


QuoteOriginally posted by photodesignch Quote
It's good that SONY did indicated that anything under ISO100 is actually "simulated ISO". At least they are honest about it.
Why are you capitalizing "Sony"? Sony is just a noun, it doesn't stand for anything. Do I need to add this grammatical faux pas to my list of rage-inducing brand language, like people who capitalize "Mac" and "Leica"? Not to sound like a troll, just a pet peeve.
01-05-2015, 08:35 PM   #25
Pentaxian
Digitalis's Avatar

Join Date: Mar 2009
Location: Adelaide.
Posts: 8,797
Some of the early Nikon based Kodak SLR cameras had ISOs as low as 6.
01-05-2015, 10:05 PM   #26
Pentaxian
ScooterMaxi Jim's Avatar

Join Date: Jul 2009
Location: Minnesota
Posts: 1,501
QuoteOriginally posted by Omestes Quote
Is there anything to back this up? I generally keep all the correction settings off, hoping for a more unprocessed image. What exactly does highlight correction do, either hardware or software wise?
As far as I know, those of us who have tested it out and compared the results extensively - overwhelmingly tend to use HC in Auto mode; goes into effect only if you need it. Some people don't care if they have blown highlights. I can't afford that, especially knowing that HC with its excellent roll-off design actually looks more natural in high-contrast scenes. Frankly, all images are processed. How they are processed varies depending on how the processor is set up. Nikon 7100 achieves greater dynamic range than than the K-3 (with the same sensor) at standard settings by underexposing and recovering highlights in almost the same fashion as the K-3 with HC activated. That doesn't make the Pentax suddenly "more processed" compared to Nikon
01-05-2015, 11:44 PM   #27
Site Supporter
RGlasel's Avatar

Join Date: Sep 2013
Location: Saskatoon
Photos: Gallery | Albums
Posts: 1,345
QuoteOriginally posted by vonBaloney Quote
much less limiting counting device
Bottom line is that a CMOS camera sensor doesn't count individual photons while the shutter is open. It takes many photons hitting each individual photosite at the same time to transfer enough energy to generate a measurable electrical current. Voltage is the pressure, not the volume, of that generated current, and by itself, that voltage is too small to measured at the level of light intensity we want to be able to take photographs at. So a base current is applied to the sensor and any increases in voltage from the photons hitting a specific photosite is measured instead. Voltage is a continuous value, it doesn't accumulate, but the camera's processor can calculate the volume of light energy hitting the photosite by multiplying voltage by the length of time the shutter is open.
QuoteOriginally posted by vonBaloney Quote
How can it get a signal from the metering system to tell how many photons are going to arrive if they haven't arrived yet?
The camera's light meter is on when the camera is turned on, not just when the shutter is open. It also takes voltage readings from a larger area than individual photosites on the camera sensor. It doesn't have to intercept the photons that hit the camera sensor, it just has to measure the average intensity of light coming through the lens.
QuoteOriginally posted by vonBaloney Quote
If there was no native ISO, then all ISOs would be the same (or could be the same), and you'd have equal noise across the range
Which is something like the ISOless argument in the earlier thread I linked to. There are limits, but signal to noise tests suggested that the K-5 was ISOless from ISO 80 to 1600. Within that range, there is no benefit to manipulating shutter speed or aperture in order to take a picture at the "best" ISO value. My argument is that even if there is a "best" ISO value for a particular camera, that value is determined by the electronics that adjust the sensor's sensitivity, not by the sensor itself. Different cameras could have different "best" ISO values with the same sensor.

I don't want to restrict myself to only taking pictures at my camera's "best" ISO setting. I want to select aperture and shutter speed before I select ISO, and I'll put up with some noise, if the alternative is to not take the picture at all. To the question in the original post, "would the images be even better at 50 or 25 ISO if that were an option," the answer is no, because the camera manufacturer has access to better information about the characteristics of the sensors they put in their cameras than we do, and presumably they select minimum and maximum ISO settings where the signal from the light hitting the sensor can still be reliably distinguished from the noise generated by random variations in the sensor's electronics. Different limits would require a different design and/or a different sensor.
01-05-2015, 11:58 PM   #28
Senior Member
Omestes's Avatar

Join Date: Nov 2011
Location: Phoenix, AZ
Photos: Gallery | Albums
Posts: 207
QuoteOriginally posted by ScooterMaxi Jim Quote
As far as I know, those of us who have tested it out and compared the results extensively - overwhelmingly tend to use HC in Auto mode; goes into effect only if you need it. Some people don't care if they have blown highlights. I can't afford that, especially knowing that HC with its excellent roll-off design actually looks more natural in high-contrast scenes. Frankly, all images are processed. How they are processed varies depending on how the processor is set up. Nikon 7100 achieves greater dynamic range than than the K-3 (with the same sensor) at standard settings by underexposing and recovering highlights in almost the same fashion as the K-3 with HC activated. That doesn't make the Pentax suddenly "more processed" compared to Nikon
QuoteOriginally posted by ScooterMaxi Jim Quote
...use HC in Auto mode...
Why just Auto Mode? I barely shoot in Auto, so would it make a real difference in M, P, or Av? How does it compare to just sliding back the EV by a stop or so, which I generally do for shots that risk blown highlights?

I pretty much default to plain RAW, with no options on, natural color, center focus, center weighted metering, and auto ISO between 80 and 400, and SR off (mostly). I'm very open to being blinded by habit, and missing some settings that would make things better/easier. Sadly I still don't really feel that I've mastered much on the software side of my cameras. Heck, I just learned that I don't have to turn off my camera constantly today... Which makes me feel a bit daft. Happy, perhaps, since it is nice to realize that I am capable of learning strikingly, blatantly obvious things even after 5 years of shooting Pentax.
01-06-2015, 06:03 AM   #29
Pentaxian




Join Date: Apr 2007
Location: Sweden
Posts: 1,940
QuoteOriginally posted by ScooterMaxi Jim Quote
As far as I know, those of us who have tested it out and compared the results extensively - overwhelmingly tend to use HC in Auto mode; goes into effect only if you need it. Some people don't care if they have blown highlights. I can't afford that, especially knowing that HC with its excellent roll-off design actually looks more natural in high-contrast scenes.
Does that mean you are shooting jpeg?
Otherwise HC only use one step lower iso then stated and thus you get an underexposed raw picture (with highlights preserved) which you then have to push yourself in a raw converter. (unless you are using Pentax own converter, I guess). At least that is how I thought it worked on the K5.
01-06-2015, 09:19 AM   #30
Pentaxian
ScooterMaxi Jim's Avatar

Join Date: Jul 2009
Location: Minnesota
Posts: 1,501
Guys, I've been shooting mostly RAW for more than a decade. On occasion, I get hired to shoot on deadline so I might have to submit JPEGs.

In any event, I generally shoot RAW+Jpeg. Some RAW converters pick up the HC code - I use Capture One v. 8, and LR v.5 - both pick up the code. Converters that don't pick up the code will show the RAW image underexposed, and no highlight roll-off. Coming from my days in medium format film shooting, I really notice that the modern CMOS sensors are rather poor at showing a nice S-curve, but all that can be compensated for in the RAW processing. In my opinion, the curve applied by Pentax in HC mode is excellent. Again, it pretty much matches the Nikon curve in standard shooting (comparing the K-3 and the 7100). Frankly, the K-3 applies "Auto" HC correction too conservatively - you'll get considerable blown specular highlights before the camera triggers the compensation. The Auto setting is a bit better - quicker to trigger - on some of the older bodies, such as the K-30 and K-01. In any event, yes, Highlight Compensation most certainly has an impact on RAW images; Shadow Compensation does not.

Omestes, I understand the desire to set everything yourself, not letting the camera think for you - avoiding unnatural shooting outcomes (I also prefer the Natural setting). Ultimately, the camera is a computer - and it is making choices for you even when you control certain things manually - it will never be a K1000! But seek out that desire for "learning strikingly." One of the best settings on the Pentax cameras is TAv - you dial in the shutter speed and aperture, and let the camera find the right ISO for good exposures. I think 400 is a bit extreme for limiting the auto ISO, but I shoot a lot in very low light. I don't see much downside to using SR, except it should be avoided when on the tripod and when panning. I get about 2-3 stops of improvement, allowing me to use lower ISO, or avoid setting up the tripod.

As for "Auto HC," it just means that it will trigger the compensation only when the dynamic range is too great for the sensor to handle. Of course, you can decide to turn it on or off based on your perception of the scene. Overall, I trust the camera to do it (although I sometimes add a bit extra under-exposure on the K-3 if the scene is extreme; your K-5 series camera with its great dynamic range might be better than the K-3 in that regard). Others insist on guessing underexposure and adjusting the curve in the RAW converter. Yes, you can do that - but you risk not getting it right, and you add a lot of processing time. In my experience, the camera is at least as good at making the right call as I am - and it saves me time in post. I say that knowing I'm quite skilled over the many years I've been using two of the better RAW converters on the market.

Last edited by ScooterMaxi Jim; 01-06-2015 at 09:23 AM. Reason: clearer wording
Reply

Bookmarks
  • Submit Thread to Facebook Facebook
  • Submit Thread to Twitter Twitter
  • Submit Thread to Digg Digg
Tags - Make this thread easier to find by adding keywords to it!
camera, cameras, dslr, dunno, hardware, increase, iso, low iso sensor mimimum, magic, offer, pentax, photography, tests
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
is ISO the same for all sensor sizes? dstructor Digital Processing, Software, and Printing 13 03-07-2013 09:14 AM
K-x minimum ISO rating ranmar850 Troubleshooting and Beginner Help 3 12-16-2012 06:57 PM
Aperture + *minimum* speed -> iso ?? FixB Pentax DSLR Discussion 6 04-18-2011 06:58 AM
K-5 ISO Sensor Variation? illdefined Pentax K-5 6 04-01-2011 10:33 PM
Standard minimum ISO 400 on K-x tebitan Pentax DSLR Discussion 8 12-12-2009 12:26 PM



All times are GMT -7. The time now is 11:07 AM. | See also: NikonForums.com, CanonForums.com part of our network of photo forums!
  • Red (Default)
  • Green
  • Gray
  • Dark
  • Dark Yellow
  • Dark Blue
  • Old Red
  • Old Green
  • Old Gray
  • Dial-Up Style
Hello! It's great to see you back on the forum! Have you considered joining the community?
register
Creating a FREE ACCOUNT takes under a minute, removes ads, and lets you post! [Dismiss]
Top